GASTROSTOMY TUBE SUPPLEMENTATION FOR HIV-INFECTED CHILDREN

Objective. Malnutrition is common in pediatric human immunodeficiency virus (HIV) infection, and little is known of effective nutritional in terventions. We sought to determine whether enteral supplementation wi th gastrostomy tube feedings would provide improvements in weight, hei ght, body composition, immune parameters, morbidity, and mortality. Me thods. We collected clinical data on 23 HIV-infected children who were fed chronically by gastrostomy tube. The main outcome measures includ ed weight, height, triceps skinfold thickness (TSF), arm-muscle circum ference (AMC), hospital days, caloric intake, and CD4-positive T-lymph ocyte count. Each of these parameters was measured or evaluated at fou r points: 6 months before nasogastric tube feeding, at the time nasoga stric tube feeding was initiated, at the time gastrostomy tube feeding was initiated, and 6 months after gastrostomy tube feedings began. Re sults. Weight z score [-2.1 (0.14) to -1.58 (0.14)] and weight-for-hei ght z score [-0.98 (0.16) to -0.15 (0.17)] improved with gastrostomy t ube feedings. There was a trend toward improvement in weight z score w ith nasogastric tube feedings. Caloric intakes increased progressively with nasogastric and gastrostomy tube feedings. No improvement in hei ght, TSF, AMC, hospital days, or CD4 counts was seen in the follow-up period. However, children who had the greatest increase in weight had the most improvement in fat stores (TSF) (r = .65, P = .002) and a dec rease in hospital days after the gastrostomy tube was placed (r = - .4 8, P = .025). Higher age-adjusted CD4 counts and lower weight-for-heig ht z scores at the time of enteral supplementation were significant pr edictors of a positive response to gastrostomy tube feedings (r = .85, P = .0001). Children who responded favorably had a 2.8-fold reduction in the risk of dying for every positive unit change in weight z score (P = .005). Conclusion. Gastrostomy tube supplementation for HIV-infe cted children can improve weight and fat mass when other oral methods fail. Weight gain is coincident with greater caloric intakes. HIV-infe cted children with higher CD4 counts and lower weight-for-height z sco res are likely to respond favorably to gastrostomy tube feedings. Earl y nutritional intervention is indicated for HIV-infected children.